No abstract

Steam ingestion into a gas turbine engine compression system can have effects on performance and operability. It is important to be able to quantify these effects. This paper describes a preliminary technique developed to model steam ingestion in a gas turbine engine compression system using a loosely coupled technique of a one-dimensional (1-D) vaporization code and a 1-D meanline compressor performance code. The results are presented for two separate parametric studies: first, the effects of the quantity of steam; and second, the effects of the quality of steam. The analytical results are evaluated against the observed industry standards. These results show that steam ingestion does have an effect on compression system performance and, potentially, operability. The effects can be seen in an increase in incidence from the front to the back blades of the machine, which results in stage rematching. These effects are caused mainly as a result of heat transfer during vaporization of the liquid water present in steam with quality less than 100 percent. The effects of fluid property changes (gas constant and specific heat ratio), caused by the addition of the vaporized water, are secondary.

A modeling and simulation technique has been developed to use a one-dimensional (1-D) multiphase code and a 1-D compressor meanline code to investigate the effects of water ingestion. The multiphase code primarily accounts for the heat transfer associated with the phase change of water, and the meanline code uses the heat transfer and gas properties to model the flow properties through a compression system. Validation of the combined codes is performed with fogging water experimental data along with cycle code information from a ground-based Industrial Gas Turbine, FT8. A parametric study is then conducted with a modern fan and core compressor to determine the performance trend changes caused by the ingestion of liquid and/or vapor water such as might be present during steam ingestion from a carrier-lauched aircraft.

Theoretical estimates are presented for the impulse generated by the collapse of a steam bubble trapped by a column of subcooled liquid in a pipe. Predictions are compared with experimental results obtained in a laboratory scale test system. The excitation of pipe responses by the transmission of the water hammer impulse through pipe bends is discussed and predictions are presented for two simple piping configurations. Experimentally measured pipe deflections excited by a steam-generated water hammer are compared with the predictions.